Ethnolects. Where language contact, language acquisition and dialect variation meet FRANS HINSKENS Meertens Instituut (KNAW) & Vrije Universiteit Amsterdam 1. Ethnolect. Concept, older scholarship and approaches1 Ethnolects are varieties of a language (usually the dominant language) which originated in a specific ethnic or cultural group. This preliminary definition, which will be refined below and in sections 3.1 and 3.3, is related to Thomason & Kaufman’s (1988) conception of ethnolects as products of language shift. Important ‘early’ work includes Lavandera’s (1976) description of Cocoliche, a variety of South-American Spanish which developed in the harbor area of Buenos Aires out of mixed L2 Spanish of Italian immigrants; it was named after the originally Italian actor Antonio Cocoliche who, during a performance, unintentionally slipped back into his ‘semi-Spanish’. Cocoliche Spanish is a highly variable system, which can be closer to certain Italian dialects or to the Spanish of the first generations of Italians living in the Buenos Aires harbor area. Garland Bills (1976) described an English vernacular variety has emerged among members of the Chicano community. In the southwestern part of the USA. Bills (1976) discusses the question if this variety should be considered as the result of Spanish interference in the L2 English or rather as a discrete variety of English. He argues for the status of a discrete variety (Vernacular Chicano English, henceforth VCE) for three reasons: 1) the features of VCE cannot be predicted on the basis of a contrastive analysis of Spanish and English, 2) VCE is variable, but the variability is structured and sociolinguistically conditioned, rather than in terms of the English skills of its speakers, 3) "VCE is not a [...] transitional phenomenon", as it is also used by Chicanos who do not speak Spanish. Carlock & Wölck (1981) report on the Buffalo ethnolects perception experiments, for they played fragments of the speech of inhabitants of the neighborhood West Side, an originally Italian neighborhood. Older listeners identified it as Italian English, but the younger the listeners, the more often the speech fragments were associated with the West Side – as the variety had spread to members of other groups in the neighborhood. A similar development has been discussed (Chambers 2003), who described how speech characteristics of the English of ethnic Italians in Toronto (e.g. the realization of ‘sandwich’ als sa[ŋ]wich) seem to spread to the ethnic Greeks living in the same neighborhood, East End. Danesi (1985: 118) defined an ethnolect as "the variety of a language that results when speakers of different ethnolinguistic backgrounds attempt to speak the dominant language (e.g. 'Chicano English')". In Danesi's view, ethnolects are hence products of language shift à la Thomason & Kaufman 1988. In this view, e.g. Hiberno and Scottish English also count as (former) ethnolects. Labov & Harris (1986) showed how the local varieties of English spoken by black and white inhabitants of Philadelphia are gradually diverging because the 1 Much of this paper is based on Hinskens & Muysken 2007, Hinskens 2011, Muysken 2013 [in Lg & SP:D], Van Meel et al. 2013, 2014, Hinskens 2014, 2015 and Hinskens et al. in press. _____________________________________________ In I. Kappa & M. Tzakosta (eds), 2019. Proceedings of the 7th Ιnternational Conference on Modern Greek Dialects and Linguistic Theory, 1-27. Patras: University of Patras. 2| FRANS HINSKENS black, most of whom are cut off from the ‘mainstream’, living in other neighborhoods in their own social networks, hardly participate in sound changes such as the fronting of the /ɑu/ diphthong (in items such as ‘house’). Both groups appeared to be aware of this mutual divergence, which may lead to the emergence of separate ethnolects. Kotsinas (1988) sketched how in Rynkeby and other suburbs of Stockholm peer groups of adolescent members of ethnic minority groups developed their own multilingually mixed substandard varieties of Swedish for in-group use. In a publication from 1996, the Dutch linguist Backus explicitly pled for the inclusion of ethnicity as a sociolinguistic factor in studies of language variation. Clyne was one of the scholars who introduced the notions of ethnolect which he defined as “varieties of a language that mark speakers as members of ethnic groups who originally used another language or distinctive variety” (2000: 86) and multi-ethnolect. In Androutsopoulos' (2001b: 322) view, an ethnolect is "a variety of the majority language (or 'host language') which is used by and regarded as a vernacular for speakers of a particular ethnic descent and is marked by certain contact phenomena". Language shift and incomplete second language acquisition play a minor role in Androutsopoulos' conception. The author (2001a,b) paid ample attention to the role of the mass media in the dissemination of (what is perceived as) ethnic features. This is an important link to Auer’s 2003 paper which introduces the three-way distinction primary - secondary - tertiary ethnolect (à la Coseriu’s 1980 tripartition distinction primary - secondary - tertiary ethnolect). Whereas secondary ethnolect is the transformation of the variety by media genres such as comedy or film,2 tertiary ethnolect is used by adolescents without direct knowledge of the primary ethnolect. Ethnolects have a relatively high ‘kissing number’. The ‘kissing number’ is a concept from nano-physics and it refers to a property of e.g. an atom: the number of neighboring atoms (which is high in crystalline structures). The kissing number of the ethnolect concept is 7, as work on ethnolects is relevant to the study of language contact and bilingualism, to quantitative sociolinguistics, interpretive sociolinguistics, ethnographically enriched sociolinguistics, historical linguistics, and dialectology. In general, there are two distinct approaches to the study of ethnolects: the language centered and the ethnographic approach. Whereas the ethnographic approach conceives language systems as infinite resources from which speakers may freely choose to shape their identity, the language centered approach tries to disentangle the laws, generalisations and restrictions on these resources. The language centered approach typically stands out by • the use of terminology such as ‘ethnolect’, ‘multi-ethnolect’ and ‘multicultural variety’, • ‘etic’, ‘objective’ definitions of ethnicity (language, race, descent), • quantitative methodology (often in the Labovian tradition), • focus on form, structure and the distribution of variation, • a macro-social angle. The ethnographic approach typically stands out by • the use of terminology such as ‘style’ and ‘(pan-)ethnic style’,3 • ‘emic’, ‘subjective’ definitions of ethnicity (social construction; perception),4 • attention for both reactive and initiative uses of linguistic and paralinguistic features, • interpretive methods, • focus on social meaning (‘indexicality’, ‘indexical fields’) and its fluidity, • a micro-social perspective. 2 Or, one might add, the theatre, as in the case of Cocoliche (above). 3 On concepts and terminology, see Kern (2011: 4-10). 4 Cf. Fought (2013). Ethnolects. Where language contact, language acquisition and dialect variation meet | 3 2. International perspective In Europe, important work has been done in Scandinavia – to begin with Kotsinas’ (1988) work on ‘Rinkeby Svenska’, sketched above. Fraurud & Boyd (2006) deconstruct the concept of the native speaker on the basis of data of 222 speakers in the contemporary multilingual urban settings in the Swedish cities of Göteborg, Malmö and Stockholm. Bodén (2005) focuses on features5 of the variety of Swedish spoken in Rosengård, a suburb of Malmö. Quist (2008) zooms in on sociolinguistic developments in younger varieties of Danish as spoken in Kopenhagen, which make her introduce the concept of the multi-ethnolect. For the German language area, there is a decades old sociolinguistic tradition of the study of Gastarbeiterdeutsch (roughly: immigrant workers’ German) initially as a second language and later also as a variety which is colored by the heritage language. In the German situation this is mainly Turkish. Work on what used to be called ‘Kanaksprak’ (a derogatory label) has been done by Deppermann (2007), Kern, Selting (2011) and Wiese6 (2009, 2013), focusing on the Berlin situation, Keim (2002) for Mannheim, and Auer and colleagues (Hamburg; Freiburg i.Br.). Much of the work on ethnolectal and related variation in French seems to concentrate on lexical aspects. Among the noteworthy studies are Jamin et al. (2006) on variation (mostly in the realization of specific cosonants) in the French spoken in suburbs of Marseille, Grenoble and Paris. Fagyal & Stewart (2011) study developments in certain Parisian banlieues and zoom in on phenomena such as phrase-final intonation movements. The developments in the UK are strongly determined by the consequences of postcolonial migration flows – mainly from south Asia and the Caribean region. ‘Glaswasian’, i.e. Glasgow Asian colored varieties of English have been studied by Stuart Smith et al. (2011) and the English of Pakistani and Black Caribean groups in Birmingham by Kahn (2006 - below). Cheshire et al. (2011) zoom in on the complex multifaceted developments in Modern London English. For the range of sociolinguistic situations in Africa, where "multilingualism is perceived as a normality, rather than as a special case" (Wiese 2016: section 3.1), the boundaries